Recombinomics | Elegant Evolution

Home Founder What's New In The News Contact Us

Paradigm Shift

Viral Evolution

Intervention Monitoring

Vaccine Screening

Vaccine Development

Expression Profiling

Drug Discovery

Custom Therapies


State Secrets in H5N1 Qinghai Study Published in Nature?

Recombinomics Commentary

July 13, 2005

Hong Kong University scientist Guan Yi, who led a recently published study of migratory birds struck down with avian flu in April in Qinghai, said he had been accused by mainland authorities of stealing state secrets after the study was published.

Guan was quoted in Chinese-language media Monday as saying that he had received a warning from agricultural authorities who accused him of exaggerating the extent of the outbreak and ''leaking state secrets.''

The comments above seem to indicate China considers the widespread presence of H5N1 throughout China in 2005 to be a "state secret".  Last week sequences of 4 isolates from Qingahi lake were published within a day of each other in Nature and Science.  Yi Guan lead the Hong Kong and Shantou groups publishing in Nature , while the Science paper came from a group in Beijing.  Bases on accession numbers at GenBank, the Hong Kong group submitted sequences (accession numbers DQ95612-DQ95771) prior to the Beijing group's submissions (accession numbers DQ100542-DQ100573), and both groups listed four Qingahi sequences in the dendrograms (isolates from 2 bar headed geese, and 1 each from great black headed gull and brown headed gull). 

The descriptions of these isolates in the text and figures sounds virtually identical (the actual sequences have not been made public, but will be available shortly). Thus, for the 4 sequences from Qinhai, the general descriptions are the same (all 8 isolates have E627K in PB2 and all are reassortants with three genes most closely related to a 2003 chicken isolate from Shantou and five genes most closely related to a 2004 peregrine falcon sequence from Hong Kong).

However, based on accession numbers, Beijing submitted 32 sequences to GenBank while Hong Kong submitted 160.  Assuming the sequences represent 8 genes per isolate, Beijing just submitted the sequence of the 4 Qinghai isolates, while Hong Kong submitted sequences from 20 isolates, which were described in the supplemental figure legend as follows:

"To determine the genetic relationship of this virus to other H5N1 viruses, 20 of 33 H5N1 influenza viruses isolated at this outbreak and 8 other viruses isolated from poultry market surveillance from Fujian, Guangdong, Hunan and Yunnan Provinces during 2005 were partially sequenced for each of the eight gene segments."

Moreover, the figure in the Nature paper included placement of two 2005 sequences from Shantou and well as one from Hunan and Yunnan indicating these 4 were among the 8 islates outside of Qinghai.

The Nature paper noted that for the MA gene, the sequences from the four Qinghai birds were virtually identical to the two 2005 isolates from Shantou (a duck and chicken) and concluded that the Qinghai H5N1 came from Shantou province.  However, this conclusion assumed that there was no H5N1 in India, the site of origin for many of the bar headed geese which winter in India and migrate to Qingahi Lake in May and June.  Although India has denied H5N1 is in either people or birds, three poultry workers had H5N1 antibodies in serum collected from them in 2002 and India had a large die off of pigeons in 2004, which were claimed to be H5N1 negative, but the cause of the die-off was not stated in the report to OIE denying H5N1 in India.  Moreover, now Indonesia is indicating that three suspected bird flu cases in one family, including two deaths, may be linked to the familiy's trip to India and Hong Kong.  (Sequencing of those isolates would provide strong indicators because H5N1 sequences from Indonesia from 2003 and 2004 are publicly available, and the sequence include regional markers exclusive to Indonesia.)

Since earlier Shantou isolates also are related to Qinghai and contain several polymorphisms found in migratory bird H5 sequences, the possibility that the Shantou birds were infected by virus that originated in India and then transferred to birds heading for Shantou province from Qinghai Lake cannot be excluded, but this is a matter of interpretation of the data, which also assumes India's denials of H5N1 are accurate.

However, the level of unreported H5N1 in Asia appears to be significant.  In India, the H5N1 positive poultry workers were said to have not traveled outside of India, strongly suggesting unreported H5n1 has been in India since at least 2002.  Moreover, the real "state secret' in the Nature report is the isolation of H5N1 from at least four provinces in China in 2005.  High Pathogenic Avian Influenza (HPAI) is a reportable disease and the H5N1 sequences of virtually all isolates throughout Asia have had an HPAI sequence at the HA cleavage site (which is considered diagnostic for HPAI), so the sequences would provide conclusive evidence that H5N1 was widely circulating in China in 2005, but nor reported.  The first H5N1 report to OIE for 2005 was on the H5N1 in Qinghai on May 21 followed by two reports on H5N1 in Xinjiang province.

Thus, the publication of the 2005 H5N1 sequences from Fujian, Hunan, Yunnan, and Shantou province indicates China failed to report associated outbreaks.  To deny the presence of the H5N1 in these provinces would require calling the sequences a lab artifact or contaminant, which was a tactic used by South Korea and WHO to deny that the WSN/33 sequences in swine in Korea were real.

However, the story is in the sequence, and an evolving WSN/33 in Korean swine and the 2005 sequences in eastern China clearly indicate the data are quite real, with or without press releases on lab artifacts or errors.

Media sources

Home | Founder | What's New | In The News | Contact Us

© 2005 Recombinomics.  All rights reserved.